Central limit theorems for the outputs of fully convolutional neural networks with time series input
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866908925113139200 |
|---|---|
| author | Betken, Annika Micali, Giorgio Schmidt-Hieber, Johannes |
| author_facet | Betken, Annika Micali, Giorgio Schmidt-Hieber, Johannes |
| contents | Deep learning is widely deployed for time series learning tasks such as classification and forecasting. Despite the empirical successes, only little theory has been developed so far in the time series context. In this work, we prove that if the network inputs are generated from short-range dependent linear processes, the outputs of fully convolutional neural networks (FCNs) with global average pooling (GAP) are asymptotically Gaussian and the limit is attained if the length of the observed time series tends to infinity. The proof leverages existing tools from the theoretical time series literature. Based on our theory, we propose a generalization of the GAP layer by considering a global weighted pooling step with slowly varying, learnable coefficients. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_29612 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Central limit theorems for the outputs of fully convolutional neural networks with time series input Betken, Annika Micali, Giorgio Schmidt-Hieber, Johannes Methodology Machine Learning Deep learning is widely deployed for time series learning tasks such as classification and forecasting. Despite the empirical successes, only little theory has been developed so far in the time series context. In this work, we prove that if the network inputs are generated from short-range dependent linear processes, the outputs of fully convolutional neural networks (FCNs) with global average pooling (GAP) are asymptotically Gaussian and the limit is attained if the length of the observed time series tends to infinity. The proof leverages existing tools from the theoretical time series literature. Based on our theory, we propose a generalization of the GAP layer by considering a global weighted pooling step with slowly varying, learnable coefficients. |
| title | Central limit theorems for the outputs of fully convolutional neural networks with time series input |
| topic | Methodology Machine Learning |
| url | https://arxiv.org/abs/2603.29612 |